PDA

View Full Version : Unusual autosomal matches between Ancient DNA and modern British kits



David Mc
03-15-2015, 09:17 AM
In the "Do you match Yamnaya or EHG samples" thread, a number of us decided to see what matches Gedmatch would show between us and some archaic samples they have mapped out.

I am including my own results below, but I would love to see how others of British (or even northwestern European in general) background map out here.

I have three questions in particular:

1) Why would I have such strong matches with samples like these:

F999933 (BR2, Hungary, 3.2ky) Largest segment = 4.9 cM Total of segments > 1 cM = 877.6 cM
F999916 (LBK, Stuttgart, 7ky) Largest segment = 5.4 cM Total of segments > 1 cM = 835.3 cM
F999937 (NE1, Hungary, 7.2ky) Largest segment = 4.2 cM Total of segments > 1 cM = 814.5 cM
F999918 (Loschbour, Lux. 8ky) Largest segment = 4.4 cM Total of segments > 1 cM = 674.8 cM
F999935 (Ust-Ishim, Siberia, 45ky) Largest segment = 4.7 cM Total of segments > 1 cM = 632.0 cM

... and such weak matches with the Hinxton samples (c.f. below), which are from the British Isles and significantly more recent? Surely this isn't what we would expect?

2) How could I possibly get 307,5 cM with Clovis (remembering I am entirely British), when the Hinxton samples only give me 6.2 cM, 2.4 cM, 2.9 cM, and 0.0 cM, respectively?

3) Why are the Hungarian matches so strong? It suggests an historic or prehistoric movement of people through the Hungarian plains who contributed a significant amount of material into our Western European gene pool. What cultures would the original samples represent?

My info, and the method we used are found below.


Comparing kit F216031 (Ethnicity: British) with Archaic Autosomal Samples from Gedmatch Database
Using minimum threshold = 200 SNPs; mismatch-bunching limit = 25 SNPs; minimum segment cM to be included in total = 1.0 cM, listed chronologically:

F999902 (Altai N., Siberia, 50ky) Largest segment = 1.2 cM Total of segments > 1 cM = 2.4 cM
FF999935 (Ust-Ishim, Siberia, 45ky) Largest segment = 4.7 cM Total of segments > 1 cM = 632.0 cM
999936 (Kostenki14, Russia, 37ky) Largest segment = 4.4 cM Total of segments > 1 cM = 73.8 cM
F999903 (Denisova, Siberia, 30ky) Largest segment = 0.0 cM Total of segments > 1 cM = 0.0 cM
F999914 (Malta, Siberia, 24ky) Largest segment = 3.3 cM Total of segments > 1 cM = 18.4 cM
F999919 (Clovis, Montana, 12.5ky) Largest segment = 5.4 cM Total of segments > 1 cM = 307.5 cM
F999918 (Loschbour, Lux. 8ky) Largest segment = 4.4 cM Total of segments > 1 cM = 674.8 cM
M652848 (EHG_Oleniy_Ostrov I0061, 7.5ky?) Largest segment = 8.4 cM Total of segments > 1 cM = 8.4 cM
F999931 (KO1, Hungary, 7.7ky) Largest segment = 3.7 cM Total of segments > 1 cM = 20.2 cM
F999937 (NE1, Hungary, 7.2ky) Largest segment = 4.2 cM Total of segments > 1 cM = 814.5 cM
F999932 (NE6, Hungary, 7.1ky) Largest segment = 2.5 cM Total of segments > 1 cM = 9.6 cM
F999927 (NE5, Hungary, 7.1ky) Largest segment = 2.5 cM Total of segments > 1 cM = 18.8 cM
F999917 (Motala-12, Sweden, 7ky) Largest segment = 3.0 cM Total of segments > 1 cM = 69.6 cM
F999915 ( (La Brana, Spain, 7ky) Largest segment = 2.7 cM Total of segments > 1 cM = 40.6 cM
F999916 (LBK, Stuttgart, 7ky) Largest segment = 5.4 cM Total of segments > 1 cM = 835.3 cM
F999928 (NE7, Hungary, 6.4ky) Largest segment = 3.7 cM Total of segments > 1 cM = 20.2 cM
M303249 (Esperstedt-MN I0172, 5.3ky?) Largest segment = 6.6 cM Total of segments > 1 cM = 45.3 cM
M020637 (Yamnaya_Sok_River I0443, 5.3ky?) Largest segment = 6.6 cM Total of segments > 1 cM = 36.1 cM
F999924 (Ajvide58, Sweden, 5ky) Largest segment = 2.9 cM Total of segments > 1 cM = 10.0 cM
F999934 (Gokhem2, Sweden, 5ky) Largest segment = 2.4 cM Total of segments > 1 cM = 7.5 cM
M951285 (Yamnaya I0231, 4.9ky) Largest segment = 10.4 cM Total of segments > 1 cM = 41.1 cM
F999930 (CO1, Hungary, 4.8ky) Largest segment = 3.8 cM Total of segments > 1 cM = 27.4 cM
M117132 (Bell-Beaker I0112, 4.4ky?) Largest segment = 7.5 cM Total of segments > 1 cM = 23.6 cM
F999906 (P.Eskimo,Greenland, 4ky) Largest segment = 2.2 cM Total of segments > 1 cM = 15.6 cM
F999933 (BR2, Hungary, 3.2ky) Largest segment = 4.9 cM Total of segments > 1 cM = 877.6 cM
F999929 (IR1, Hungary, 2.9 ky) Largest segment = 2.4 cM Total of segments > 1 cM = 15.9 cM
F999925 (Hinxton-4, UK, 2ky) Largest segment = 2.7 cM Total of segments > 1 cM = 6.2 cM
F999921 (Hinxton-2, UK, 1.3ky) Largest segment = 1.4 cM Total of segments > 1 cM = 1.4 cM
F999922 (Hinxton-3, UK, 1.3ky) Largest segment = 1.8 cM Total of segments > 1 cM = 2.9 cM
F999926 (Hinxton-5, UK, 1.3ky) Largest segment = 0.0 cM Total of segments > 1 cM = 0.0 cM

Shaikorth
03-15-2015, 10:09 AM
1) Why would I have such strong matches with samples like these:

F999933 (BR2, Hungary, 3.2ky) Largest segment = 4.9 cM Total of segments > 1 cM = 877.6 cM
F999916 (LBK, Stuttgart, 7ky) Largest segment = 5.4 cM Total of segments > 1 cM = 835.3 cM
F999937 (NE1, Hungary, 7.2ky) Largest segment = 4.2 cM Total of segments > 1 cM = 814.5 cM
F999918 (Loschbour, Lux. 8ky) Largest segment = 4.4 cM Total of segments > 1 cM = 674.8 cM
F999935 (Ust-Ishim, Siberia, 45ky) Largest segment = 4.7 cM Total of segments > 1 cM = 632.0 cM

... and such weak matches with the Hinxton samples (c.f. below), which are from the British Isles and significantly more recent? Surely this isn't what we would expect?

2) How could I possibly get 307,5 cM with Clovis (remembering I am entirely British), when the Hinxton samples only give me 6.2 cM, 2.4 cM, 2.9 cM, and 0.0 cM, respectively?

3) Why are the Hungarian matches so strong? It suggests an historic or prehistoric movement of people through the Hungarian plains who contributed a significant amount of material into our Western European gene pool. What cultures would the original samples represent?


The answer to all of these is related to the quality of ancient sequences. The genomes we have lots of matching with (Ust-Ishim, Clovis, BR2, LBK etc) are all high coverage which preserves IBD chunks. The Hinxtons OTOH are uniformly low coverage sequences which makes their SNP's look homozygous and thus destroys IBD. This makes them not match anyone very well even though they are much closer to modern Europeans than Loschbour and LBK, never mind Ust-Ishim. They will get reasonable results in calculators though.

David Mc
03-15-2015, 10:27 AM
Thank you, Shaikorth. I have been rather busy of late and feel like I'm falling behind on things that are probably fairly obvious to others.

tipirneni
11-26-2017, 04:32 AM
StuttGart LBK sample on GED comparison

Largest segment = 3.8 cM
Total of segments > 1 cM = 660.9 cM
444 matching segments

BR2 matching segment
Largest segment = 4.6 cM
Total of segments > 1 cM = 751.9 cM
505 matching segments

KO1, Hungary, 7.7ky
Largest segment = 3.3 cM
Total of segments > 1 cM = 70.6 cM
50 matching segments

CO1, Hungary, 4.8k
Largest segment = 2.9 cM
Total of segments > 1 cM = 34.9 cM
23 matching segments

NE1, Hungary
Largest segment = 4.0 cM
Total of segments > 1 cM = 656.2 cM
442 matching segments

Clovis, Montana, 12.5ky
Largest segment = 5.9 cM
Total of segments > 1 cM = 560.2 cM
374 matching segments

RISE 552 Russian 4ky
Largest segment = 3.5 cM
Total of segments > 1 cM = 189.5 cM
126 matching segments



Hinxton-4, UK, 2ky
Largest segment = 1.7 cM
Total of segments > 1 cM = 19.0 cM
15 matching segments

Historically big mixture with Kamboja (1500BC) mixed aryan people from BMAC/Steppe, Scythian (1500BC) from Steppe, Pallava(500CE) mixed aryan people from BMAC, Hunas (600CE) white hun from Steppe